Health & wellness
    Next Score View the next score

    Study of brain tumor cells reveals genetic diversity underlying failure of treatments

    Genetic tests have become the cutting-edge in cancer treatment, allowing doctors to choose drugs that target genetic mutations carried in a patient’s tumor. But the success of such therapies came with a quick letdown: the drugs eventually stopped working.

    Tumors turn out to be made up of cells don’t all have identical DNA, meaning that even a very effective targeted drug might leave behind a subpopulation of cancer cells that eventually kill a patient.

    A new study of the gene activity of hundreds of individual cells from five patients’ deadly brain tumors adds yet another layer of complexity, revealing an enormous diversity of which genes are flipped on and off in cells in the same tumor. The study, published online in the journal Science Thursday, is a collaboration that crosses the boundary between medicine and science.


    Surgeons from Massachusetts General Hospital gathered freshly removed tumor tissue from patients, prepped it for analysis, and then sent it across the river to the Broad Institute in Cambridge, where scientists were able to use sophisticated sequencing techniques to determine the activity of genes in each individual cell.

    Get The Weekender in your inbox:
    The Globe's top picks for what to see and do each weekend, in Boston and beyond.
    Thank you for signing up! Sign up for more newsletters here

    The results are, on one hand, depressing: they show how complex glioblastoma tumors are, comprised of many different kinds of cells. And those cells are diverse in ways that matter for treatment -- drugs that are designed to attack specific molecular targets will work on some of the cells found in a single tumor, but not on others.

    But the news isn’t all bad: the technique used in the study could be used to help understand why glioblastoma -- and other cancers -- evade treatment, and could help scientists think in a more informed way about how to design the laboratory tests that they use to screen new drugs and combinations of therapies.

    “What we found here is every tumor we looked at had cells that conform to multiple different cell types,” said Dr. Brad Bernstein, a professor of pathology at Mass. General. “So you might say this patient has glioblastoma of tumor type X, and maybe most of its cells do conform to type X, but there are others that look like type Z.”

    Scientists also found evidence that some of the cells in the tumors were more like stem cells than others, suggesting that a subpopulation of cells could be the ones that seed the tumor. However, the entire notion of cancer stem cells has been under vigorous debate, in part because different groups of scientists have been unable to consistently determine which biomolecular markers truly indicate that a cancer cell is a stem cell.


    Dr. Eric Holland, director of the human biology division at the Fred Hutchinson Cancer Research Center in Seattle, who was not involved in the work, said that the variety of cells observed in the study neatly illustrates the challenge that oncologists face with glioblastoma.

    A single brain tumor’s diverse “population of cells are going to have all sorts of reasons they are going to be resistant to almost anything we’ve tried up to this point,” Holland said. “That’s an inherent problem. Right now it explains the problem -- it gives you an idea that single agents aren’t going to be very effective.”

    Bernstein said he hopes that the information can be used to create better laboratory versions of tumors to test potential drug combinations. It also may help researchers conceive of tumors as almost like an organ with different types of cells that can grow and evolve, rather than a uniform collection of one type of cancer cell.

    He admitted that initially he found the data somewhat disheartening, but when he showed it to colleagues who see patients with the devastating brain cancer, he got the opposite reaction. The data help confirm what doctors had already intuited from years of treating patients with drug therapies that almost inevitably fail.

    “This may be to my knowledge the first study that tried to do this carefully within individual cells from human tumors and it is a bummer, because this is why cancer is so hard to cure,” said Sean Morrison, director of the Children’s Medical Center Research Institute at the University of Texas Southwestern. “It’s a different battle in every patient in some ways. And this heterogeneity is why the best ideas we often have will kill 90 percent of the cells and leave the other 10 percent behind.”

    Carolyn Y. Johnson can be reached at Follow her on Twitter @carolynyjohnson.